Synthesis of electron-accepting polymers containing phenanthra-9,10-quinone units

Article abstract of DOI:10.1039/b901853k

Polymers containing 9,10-phenanthraquinone moieties as part of an extended π-electron system have been prepared by Suzuki couplings between 2,7- or 3,6-dibromophenanthraquinone and 9,9-dioctylfluorene-2,7-diboronic acid bispinacol ester. Closely related model compounds were prepared similarly. All the products were characterised by FT-IR, ¹H NMR and UV-vis/fluorescence spectroscopy and, where relevant, mass spectrometry. Cyclic voltammetry measurements indicate that these materials display electron affinities of up to 4.0 eV. This value is one of the highest reported for conjugated polymers.

Full text of DOI:10.1039/b901853k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Synthesis of electron-accepting polymers containing
phenanthra-9,10-quinone units†
a
a
Julien E. Gautrot, Philip Hodge, Madeleine Helliwell,b James Rafteryb and Domenico Cupertinoc
*
*
Received 28th January 2009, Accepted 26th March 2009
First published as an Advance Article on the web 11th May 2009
DOI: 10.1039/b901853k
Polymers containing 9,10-phenanthraquinone moieties as part of an extended p-electron system have
been prepared by Suzuki couplings between 2,7- or 3,6-dibromophenanthraquinone and 9,9-
dioctylfluorene-2,7-diboronic acid bispinacol ester. Closely related model compounds were prepared
1
similarly. All the products were characterised by FT-IR, H NMR and UV-vis/fluorescence
spectroscopy and, where relevant, mass spectrometry. Cyclic voltammetry measurements indicate that
these materials display electron affinities of up to 4.0 eV. This value is one of the highest reported for
conjugated polymers.
in reduced forms, such as the anion-radical or the dianion that
Introduction
are anthracene derivatives, these are conjugated.
In recent years there has been great interest in polymers that have
a p-electron system along the backbone.1–3 Frequently studied
examples include polyacetylenes,4 poly(phenylene vinylene)s,5
polypyrroles6 and polythiophenes.7 Polymers that have such p-
electron systems can, for example, function as light-emitting
diodes (LEDs),5,8–10 field effect transistors (FETs),5,11–18 or key
components in photovoltaic devices.19–21 Polymers which have
the further feature that they are good electron acceptors are
potentially useful for a variety of electronic applications. Thus
One type of quinone moiety that can be fully conjugated with
far one of the more successful polymers of this type is
poly(oxobenzimidazoquinoline).22 The present paper is con-
a backbone p-electron system is that present in 9,10-phenan-
thraquinone (2). Moreover, this quinone is a more powerful
electron acceptor than 9,10-anthraquinone (1).33 Another inter-
cerned with a new type of electron-accepting material where the
acceptor is a quinone unit.
esting feature of quinone 2 is the fact that the two carbonyl
In Nature a type of molecule widely used for accepting elec-
groups are on the same ‘side’ of the molecule with the result that
trons is quinones. These, for example, play a key role in oxidative
it has a significant dipole. This is of interest because it is often
phosphorylation (derivatives of benzoquinone)23 and photosyn-
crucial for the thin films used in electronic devices to be ordered.
thesis (derivatives of benzoquinone and of naphthoquinone).24
The dipoles are likely to cause derivatives containing these units
For electronic applications, however, derivatives of 9,10-
to stack with the dipoles anti-parallel to each other. The stacking
anthraquinone (1) are a more obvious type to consider because
of the quinone units can be expected to facilitate electron
apart from accepting electrons readily they generally have
transfers between polymer chains.
a higher chemical and thermal stability than benzoquinones and
In this paper we describe the synthesis and characterisation of
naphthoquinones. A few polymers containing anthraquinone
some polymers and closely related model compounds containing
units in the backbone have been synthesised.25–32 Ideally,
9,10-phenanthraquinone moieties. In order to improve the
however, for electronic applications the moieties that accept the
solubility of the materials, the quinone monomers were co-
electrons should be an integral part of a fully conjugated back-
polymerised with a dioctylfluorene derivative. Since this work
bone p-electron system. Unfortunately the two aromatic rings in
was completed34 a group based in China has reported the
anthraquinone (1) are not fully conjugated to each other, though
synthesis and characterisation of one polymer that is closely
related to the one described here.35 However, the spectroscopic
measurements and the interpretations of these differ significantly
from ours. The analyses based on the X-ray studies are new.
aOrganic Materials Innovation Centre, School of Chemistry, University of
Manchester, Oxford Road, Manchester, UK M13 9PL. E-mail: jeg45@
cam.ac.uk; Philip.Hodge@man.ac.uk
Experimental
bSchool of Chemistry, University of Manchester, Oxford Road,
General methods
Manchester, UK M13 9PL
cAvecia, Blackley, Manchester, UK M9 8ZS
These are as given previously.36 9,9-Dioctylfluorene-2-boronic
† Electronic supplementary information (ESI) available: CCDC
reference numbers 701642 and 701643. For crystallographic data in
acid pinacol ester (3)36,37 and 9,9-dioctylfluorene-2,7-diboronic
CIF or other electronic format see DOI: 10.1039/b901853k
acid bispinacol ester (4) were synthesised as described
This journal is ª The Royal Society of Chemistry 2009
4148 | J. Mater. Chem., 2009, 19, 4148–4156

View Article Online
previously.36 DCM ¼ dichloromethane; THF ¼ tetrahydro-
furan. In SEC analyses solutions were filtered with 0.22 mm filters
prior to injection into the instrument.
gave quinone 8 asꢀyellow needles (8.53 g, 69%). It had mp 286–
1
1390, 1326, 1282, 1222, 1099, 1079, 1020, 919, 896 and 820; H
287 C (lit.,40 286 C); IR (NaCl, cmꢂ1) 1676, 1581, 1545, 1467,
ꢀ
NMR (CDCl3 and d-TFA, ppm) d 8.13 (2H, d, J ¼ 2 Hz, H-4),
8.07 (2H, d, J ¼ 8 Hz, H-1) and 7.69 (2H, dd, J ¼ 2 and 8 Hz, H-
2); 13C NMR (CDCl3 and d-TFA, ppm) d 128.0, 129.1, 132.7,
134.0, 134.1, 136.6 and 180.0; MS (EI/CI) triplet 364/366/368 g
molꢂ1, C14H6O2Br2 requires 366 g molꢂ1. By HPLC analysis the
final product was >98% pure.
Synthesis of materials
2,7-Dibromo-9,10-dihydrophenanthrene
(5).
9,10-Dihy-
drophenanthrene (6) (1.36 g, 7.54 mmol), DCM (30 mL) and iron
filings (10 mg) were placed in a round-bottom flask (100 mL)
fitted with a septum and wrapped in aluminium foil to exclude
light. The mixture was stirred vigorously and cooled to 0 ꢀC.
Bromine (2.53 g, 15.8 mmol) in DCM (30 mL) was added via the
septum over 2 h. The mixture was stirred for a further 12 h at 20
ꢀC. It was then washed with aqueous sodium thiosulfate (0.1 M,
3 ꢁ 50 mL) and water (3 ꢁ 50 mL), and dried over magnesium
sulfate. The organic solvent was evaporated off and the residue
recrystallised from ethyl acetate. This gave compound 5 as white
crystals (1.41 g, 55%), mp 160–162 ꢀC (lit.,38 168 ꢀC); IR (NaCl,
cmꢂ1) 2947, 2906, 2836, 1588, 1569, 1469, 1428, 1388, 1293, 1185,
1076, 1001, 879 and 812; 1H NMR (CDCl3, ppm) d 8.30 (2H, dd,
J ¼ 1 and 2 Hz, H-3), 7.88 (2H, d, J ¼ 2 Hz, H-4) and 7.87 (2H, d,
J ¼ 1 Hz, H-1); 13C NMR (CDCl3, ppm) d 28.8, 121.8, 125.5,
130.4, 131.4, 132.9 and 139.4; MS (EI/CI) triplet 336/338/340 g
molꢂ1, C14H10Br2 requires 338 g molꢂ1 for Br ¼ 80.
2,7-Bis(90,90-dioctylfluoren-20-yl)-9,10-phenanthraquinone (9).
2,7-Dibromophenanthraquinone (7) (161 mg, 0.44 mmol), 9,9-
dioctylfluorene-2-boronic acid pinacol ester (3) (500 mg, 0.97
mmol), and palladium[0] tetrakis(triphenylphosphine) (51 mg, 50
mmol) were placed in a round-bottom flask (3-neck, 100 mL).
THF (20 mL, degassed with argon) and aqueous sodium
carbonate (1.0 M, 5 mL, degassed with argon) were added via
a septum. The mixture was heated under reflux for 28 h and then
poured onto aqueous hydrochloric acid (0.1 M, 200 mL). The
aqueous phase was extracted with DCM (3 ꢁ 50 mL), the
extracts washed with water (3 ꢁ 50 mL), and the organic solution
dried. Evaporation of the solvent gave the crude product.
Chromatography over a column of silica with a mixture of
petroleum ether–DCM (9 : 1 v/v) gave comꢀpound 9 as a violet
solid (193 mg, 45%). It had mp (DSC) 133 C; IR (KBr, cmꢂ1
)
2,7-Dibromophenanthraquinone (7). Chromium trioxide (0.50
g, 5 mmol), acetic acid (18 mL) and water (2 mL) were placed in
a round-bottom flask equipped with a thermometer and drop-
ping funnel. The mixture was stirred magnetically and cooled in
an ice bath. A solution of 2,7-dibromo-9,10-dihydrophenan-
threne (5) (0.50 g, 1.48 mmol) in acetic acid (10 mL) was added
dropwise whilst keeping the temperature below 15 ꢀC. The dark
2926, 2854, 1676, 1596, 1466, 1451, 1353, 1316, 1154, 1000, 969,
889, 823, 739 and 719; 1H NMR (CDCl3, ppm) d 8.53 (2H, d, J ¼
2 Hz, H-1), 8.14 (2H, d, J ¼ 8 Hz, H-4), 8.05 (2H, dd, J ¼ 2 and 8
Hz, H-3), 7.86–7.61 (8H, bm, Ar-H), 7.44–7.29 (6H, bm, Ar-H),
2.10–1.92 (8H, bm, aliphatic C-H), 1.26–0.96 (40H, bm, aliphatic
C-H), 0.85–0.75 (12H, bm, aliphatic C-H) and 0.72–0.51 (8H,
bm, aliphatic C-H); 13C NMR (CDCl3, ppm) d 14.3, 22.8, 23.9,
29.4, 30.2, 32.0, 40.5, 55.6, 119.2, 120.2, 120.5, 121.2, 123.2,
124.8, 126.0, 126.9, 127.1, 128.9, 129.1, 131.4, 133.9, 134.6, 137.5,
140.6, 141.1, 141.9, 142.9, 144.3, 150.1, 151.3, 151.5 and 180.9;
ꢀ
brown mixture was stirred overnight at 20 C and then poured
over crushed ice (250 mL). The precipitate was filtered off,
washed with water (3 ꢁ 10 mL) and then with methanol (3 ꢁ 10
mL). Recrystallisation of the dried product from nitrobenzene
gave quinone 7 as orange needles (0.37 g, 68%) with mp 327–328
ꢀC (lit.,39 323 ꢀC); IR (NaCl, cmꢂ1) 1674, 1581, 1463, 1398, 1263,
1223, 1203, 1145, 1080, 1037, 905 and 832; 1H NMR (CDCl3 and
d-TFA, ppm) d 8.31 (2H, m, H-4), 7.89 (2H, s, H-1) and 7.88 (2H,
m, H-3); 13C NMR (CDCl3 and d-TFA, ppm) d 125.1, 126.2,
130.4, 131.2, 134.2, 140.7 and 179.9; MS (EI/CI) triplet 364/366/
368 g molꢂ1, C14H6O2Br2 requires 366 g molꢂ1. By HPLC anal-
ysis the final product obtained was >99.9% pure.
MS (MALDI) 987 g molꢂ1, C72H88O2 requires 985 g molꢂ1
microanalysis: calc: C, 87.7%, H, 9.0%; found: C, 87.6%, H,
9.2%; UV-vis spectrum (chloroform, nm) lmax (3/L molꢂ1 cmꢂ1
;
)
334 (26 100), 494 (2100) and 525 (2100). See text for CV results.
ꢀ
By TGA it had Tdec 276 C.
3,6-Bis(90,90-dioctylfluoren-20-yl)-9,10-phenanthraquinone (10).
Compound 10 was prepared, using a procedure similar to that
described for compound 9, from 3,6-dibromophenanthraquinone
(8) (500 mg, 1.37 mmol), 9,9-dioctylfluorene-2-boronic acid
pinacol ester (3) (1.77 g, 3.42 mmol), palladium[0] tet-
rakis(triphenylphosphine) (162 mg, 140 mmol), THF (40 mL,
degassed with argon) and aqueous sodium carbonate (1.0 M, 5
mL, degassed with argon). Chromatography of the crude
product over silica with a mixture of petroleum ether (bp 40–60
ꢀC)–DCM (9 : 1 v/v) afforded compound 10 as an amorphous
orange solid (872 mg, 88%). By TGA it had Tdec 309 ꢀC. It had IR
(KBr, cmꢂ1) 2926, 2854, 1676, 1594, 1465, 1453, 1396, 1314, 1292,
1234, 1134, 1004, 926, 885, 826 and 740; 1H NMR (CDCl3, ppm)
d 8.39 (2H, d, J ¼ 1 Hz, H-4), 8.33 (2H, d, J ¼ 8 Hz, H-1), 7.85 (2H,
d, J ¼ 8 Hz, H-2), 7.82–7.64 (8H, bm, Ar-H), 7.44–7.33 (6H, bm,
Ar-H), 2.06 (8H, bm, aliphatic C-H), 1.20–0.99 (40H, bm,
aliphatic C-H), 0.78 (12H, t, J ¼ 6.7 Hz, aliphatic C-H) and 0.69
(8H, bm, aliphatic C-H); 13C NMR (CDCl3, ppm) d 14.3, 21.4,
3,6-Dibromophenanthraquinone (8). Phenanthraquinone (2)
(7.00 g, 33.7 mmol) and acetic acid (100 mL) were charged into
a round-bottom flask (250 mL) fitted with a condenser and
a hydrogen bromide trap. The mixture was heated to reflux and
bromine (24.00 g, 134.8 mmol) was added in eight portions over 8
days (before each addition, the mixture was allowed to cool
down to 20 ꢀC to avoid bromine emissions) and the reaction
followed by HPLC. After 8 days no starting material could be
detected and the reaction had stopped. The reaction was deemed
to be complete. The mixture was poured into water (400 mL).
The yellow precipitate was filtered off, washed with aqueous
sodium thiosulfate (0.1 M, 3 ꢁ 50 mL), water (3 ꢁ 50 mL) and
methanol (3 ꢁ 50 mL). The solid remaining was recrystallised
three times from xylene and dried under vacuum for 48 h. This
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 4148–4156 | 4149

View Article Online
22.9, 24.1, 29.5, 30.3, 32.0, 40.6, 55.6, 120.4, 121.8, 122.9, 123.3,
126.6, 127.3, 128.0, 128.6, 130.0, 131.5, 136.5, 138.4, 140.4, 142.7,
corrections were applied using the program SADABS.41 The
structures were solved and refined using SHELX97.41 The non-
hydrogen atoms were refined anisotropically and the hydrogen
149.5, 151.4, 152.2 and 180.4; MS (MALDI) 989 g molꢂ1
,
C72H88O2 requires 985 g molꢂ1; microanalysis: calc: C, 87.7%, H,
9.0%; found: C, 87.8%, H, 8.7%; UV-vis spectrum (chloroform,
nm) lmax (3/L molꢂ1 cmꢂ1) 301 (57 200), and 414 (23 500). See text
for CV results.
atoms were placed in idealised positions (C–H ¼ 0.95 A) and
˚
assigned isotropic thermal parameters of 1.2 times those of the
parent atoms. Computations were carried out using the
SHELXTL program package.42
Poly(9,10-phenanthraquinone-2,7-diyl-alt-9,9-dioctylfluorene-
20,70-diyl) (11). Polymer 11 was prepared, using a procedure
similar to that described above for the preparation of compound
9, from 2,7-dibromophenanthraquinone (7) (250 mg, 0.68 mmol),
9,9-dioctylfluorene-2,7-diboronic acid bispinacol ester (4) (439
mg, 0.68 mmol), palladium[0] tetrakis(triphenylphosphine) (81
mg, 70 mmol), THF (20 mL, degassed with argon) and aqueous
sodium carbonate (1.0 M, 5 mL, degassed with argon). This
afforded polymer 11 as a yellow solid (373 mg, 92%). IR (NaCl,
cmꢂ1) 2926, 2853, 2679, 1676, 1598, 1461, 1436, 1312, 1256, 1148,
1098, 998, 971, 889, 817 and 756; 1H NMR (CDCl3, ppm) d 8.53
(2H, m, H-1), 8.13 (2H, m, H-4), 8.05 (2H, m, H-3), 7.93–7.28
(6H, bm, Ar-H), 2.12 (4H, bm, aliphatic C-H), 1.11 (20H, bm,
aliphatic C-H) and 0.79 (10H, bm, aliphatic C-H); 13C NMR
(CDCl3, ppm) d 14.1, 22.6, 23.9, 29.2, 30.0, 30.7, 31.7, 40.4, 55.7,
120.6, 121.1, 124.7, 126.0, 127.2, 128.7, 128.8, 131.2, 132.1, 134.4,
137.6, 140.9, 142.5, 152.2 and 180.6; SEC (THF, g molꢂ1) Mn
6200, Mw 16 000; (CHCl3) Mn 153 000, Mw 438 000; UV-vis
spectrum (chloroform, nm) lmax (3/L molꢂ1 cmꢂ1) 364 (43 700),
493 (3900) and 526 (4100). See text for CV results. By DSC no
Crystal data for compound 7. C14H6Br2O2, Mw 366.01,
ꢀ
triclinic, space group P1, a ¼ 7.307(2), b ¼ 7.976(2), c ¼ 11.348(2)
ꢀ
3
˚
˚
A, a ¼ 73.870(3), b ¼ 77.321(4), g ¼ 64.943(3) , V ¼ 571.5(2) A ,
Z ¼ 2, Dc ¼ 2.127 g cmꢂ3, m(MoKa) ¼ 7.079 mmꢂ1, F(000) ¼ 352,
T ¼ 100 K. Crystal dimensions were 0.3 ꢁ 0.2 ꢁ 0.01 mm. 2875
reflections measured, 1987 independent reflections (Rint ¼ 0.094),
R1 ¼ 0.084 for the 858 reflections with I > 2s(I), wR(F2) ¼ 0.185
(all data).
Crystal data for compound 8. C14H6Br2O2, Mw 366.01,
monoclinic, space group P21/n, a ¼ 11.683(6), b ¼ 7.257(4), c ¼
ꢀ
3
˚
˚
14.885(8) A, b ¼ 109.542(8) , V ¼ 1189.2(10) A , Z ¼ 4, Dc ¼
2.044 g cmꢂ3, s(MoKa) ¼ 6.805 mmꢂ1, F(000) ¼ 704, T ¼ 100 K.
Crystal dimensions were 0.2 ꢁ 0.1 ꢁ 0.01 mm. 6392 reflections
measured, 2426 independent reflections (Rint ¼ 0.068), R1 ¼ 0.053
for the 1711 reflections with I > 2s(I), wR(F2) ¼ 0.104 (all data).
Results and discussion
Synthesis of 2,7-dibromophenanthraquinone (7) and 3,6-
dibromophenanthraquinone (8)
ꢀ
thermal transitions could be observed between ꢂ50 C and 120
ꢀC. By TGA it had Tdec 303 ꢀC.
Structural isomers 7 and 8 were chosen as the starting point for
the synthesis of the present polymers and model compounds. The
syntheses are outlined in Scheme 1. Various attempts to syn-
thesise 2,7-dibromophenanthraquinone (7) by the direct bromi-
nation of phenanthraquinone (2), following procedures
described in the literature, gave, by TLC analysis, complex
mixtures.40,43 The quinone was, however, prepared successfully
by a two-step method. First, 9,10-dihydrophenanthrene (6) in
DCM at 20 ꢀC was brominated by treatment with bromine in the
presence of iron filings, then the product 5 was oxidised with
chromium trioxide in aqueous acetic acid. 3,6-Dibromoan-
thraquinone (8), on the other hand, was successfully prepared
directly by reaction of 9,10-phenanthraquinone (2) with bromine
in aqueous acetic acid.44 All the products were characterised by
FT-IR and 1H NMR spectroscopy and mass spectrometry. Both
quinones had mps that are in agreement with the literature
values.
Poly(9,10-phenanthraquinone-3,6-diyl-alt-9,9-dioctylfluorene-
20,70-diyl) (12). Polymer 12 was prepared, using a procedure
similar to that described for polymer 11, from 3,6-dibromophe-
nanthraquinone (8) (400 mg, 0.63 mmol), 9,9-dioctylfluorene-
2,7-diboronic acid bispinacol ester (4) (232 mg, 0.63 mmol),
palladium[0] tetrakis(triphenylphosphine) (44 mg, 38 mmol),
THF (20 mL, degassed with argon) and aqueous sodium
carbonate (1.0 M, 5 mL, degassed with argon). This afforded
polymer 12 as an orange powder (245 mg, 65%). It had IR (NaCl,
cmꢂ1) 2926, 2853, 1677, 1594, 1465, 1399, 1313, 1291, 1233, 1135,
925, 885, 818 and 755; 1H NMR (CDCl3, ppm) d 8.34 (4H, bm,
H-1 and H-4), 7.92 (2H, bs, H-2), 7.85–7.53 (6H, bm, Ar-H), 2.17
(4H, bm, aliphatic C-H), 1.09 (20H, bm, aliphatic C-H) and 0.86–
0.61 (10H, bm, aliphatic C-H); 13C NMR (CDCl3, ppm) d 14.3,
22.8, 24.2, 29.5, 30.3, 32.0, 40.6, 56.0, 121.1, 122.0, 122.7, 122.9,
126.9, 128.7, 130.2, 131.6, 136.3, 136.4, 139.2, 141.7, 149.2, 152.6
and 180.3; (THF, g molꢂ1) Mn 3500, Mw 5300; UV-vis spectrum
(chloroform, nm) lmax (3/L molꢂ1 cmꢂ1) 309 (37 100) and 424
(22 700). See text for CV results. DSC: no thermal transition
could be observed between 0 ꢀC and 250 ꢀC; TGA: Tdec, 379 ꢀC.
X-Ray crystal structure determination
Crystals were mounted in the inert oil fomblin (per-
fluoropolymethylisopropyl ether) in a Hamilton Cryoloop. The
data were collected on a Bruker SMART APEX diffractom-
eter,41 and the crystals were cryocooled to 100 K, using an
Oxford Cryosystems 700 Series Cryostream Cooler. Absorption
Scheme 1 Synthesis of 2,7-dibromophenanthraquinone (7) and 3,6-
dibromophenanthraquinone (8).
4150 | J. Mater. Chem., 2009, 19, 4148–4156
This journal is ª The Royal Society of Chemistry 2009

View Article Online
Fig. 1 ORTEP plots of the X-ray crystal structure of (A) 8 and (B) 7,
with ellipsoids drawn at 50% probability level.
To confirm the structures of the dibromoquinones and to
determine how the molecules pack in the crystal, single crystal
X-ray crystal structures were determined. The structures, shown
in Fig. 1A and B, indicate that the isomers obtained were indeed
the expected ones based on the mps and 1H NMR spectra.
Inspection of the overall arrangement of the molecules in the
crystals additionally shows that, as expected on the basis of the
dipole moments, the molecules are stacked with alternate mole-
cules having their dipoles anti-parallel: see Fig. 2A and B. The
Fig. 2 Packing of molecules (A) 8 and (B) 7 in the crystals, with
hydrogen atoms omitted for clarity. It is evident that in each case the
molecules stack with the dipoles of neighbouring molecules anti-parallel.
˚
perpendicular spacings between the quinone layers were 3.24 A
˚
for 7 and 3.41 A for 8. Full details of the X-ray results are given in
the ESI.†
solutions in chloroform partially decolourised when left in
sunlight for several days and solutions in THF decolourised
completely. In this latter solvent, photochemical hydrogen
abstraction by 9,10-phenanthraquinone (2) is a well-documented
phenomenon.45 Because of this, the solutions prepared for the
various measurements were stored in the dark.
Synthesis of model compounds 9 and 10
Since the polymers were to be synthesised by Suzuki couplings
between the dibromoquinones and 9,9-dioctylfluorene-2,7-
diboronic acid bispinacol ester (4), model compounds 9 and 10
were prepared by similar couplings with 9,9-dioctylfluorene-2-
boronic acid pinacol ester (3):36 see Scheme 2. The products were
Synthesis of polymers 11 and 12
1
characterised by FT-IR and H NMR spectroscopy, mass spec-
trometry and elemental analysis. Differential scanning calorim-
etry (DSC) measurements revealed that 9 melts at 133 ꢀC, while
10 is amorphous. Thus, the ‘‘linear’’ compound 9 has a greater
tendency to organise and pack regularly than the ‘‘angular’’
isomer 10. Thermogravimetric analyses (TGA)ꢀrevealed that 9
and 10 are thermally stable up to 276 and 309 C, respectively.
The ultraviolet-visible (UV-vis) spectra and the cyclic voltametry
(CV) results are discussed below together with those of other
materials.
Polymers 11 and 12 were prepared by treating dibromoquinones
7 and 8, respectively, with 9,9-dioctylfluorene-2,7-diboronic acid
bispinacol ester 4,36 under similar reaction conditions to those
used to prepare the model compounds. The syntheses are sum-
1
marised in Scheme 3. The FT-IR and H NMR spectra of the
products closely resembled those of the model compounds but,
as expected, the signals in the NMR spectra of the polymers were
broader and less well defined.
The SEC results obtained for polymer 11 using THF as the
eluent indicated that the polymer had Mn of 6200 and Mw of
16 000. The trace displayed some features in its tail that are
Both 9 and 10 were stable in the solid state, that is the state in
which materials of this general type would be used in devices, but
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 4148–4156 | 4151

View Article Online
Scheme 2 Synthesis of model compounds 9 and 10: (i) Pd(PPh3)4, Na2CO3, THF.
Scheme 3 Synthesis of polymers 11 and 12: (i) Pd(PPh3)4, Na2CO3, THF.
typical of the presence of some low molecular weight oligomers:
see Fig. 3. When, however, chloroform was used as the eluent the
trace showed just one broad featureless peak and the Mn value
was 153 000 and Mw 438 000: see Fig. 3. These observations
strongly suggest that polymer 11 does not aggregate in THF but
does aggregate in chloroform. The SEC results obtained for
polymer 12 in THF showed that it had Mn 3500 and Mw 5300.
Similar results were obtained using chloroform as the eluent.
Thus, polymer 12 does not appear to aggregate in either THF or
chloroform. These observations are consistent with the fact that
polymer 11 has a more ‘‘linear’’ structure than polymer 12, and
with the fact that the ‘‘linear’’ model compound 9 is crystalline,
whilst the ‘‘angular’’ model compound 10 is amorphous. Further
evidence of aggregation phenomena is given below in the
discussion of the UV-vis and fluorescence spectra of these
materials.
that polymers 11 and 12 are thermally stable up to 303 and 379
ꢀC, respectively.
After the above syntheses had been completed a research
group based in China reported the synthesis of polymer 13.35
This only differs from polymer 11 in the length of the alkyl side
chains: polymer 13 has n-hexyl chains whereas polymer 11 has n-
octyl chains. However, as discussed below, the UV spectroscopic
data and the CV data reported for polymer 13 and some of the
interpretations of these data differ from the data we report for
polymer 11.
Photophysical properties
The UV-vis data obtained for compounds 2, 9 and 10 and
polymers 11–13 are summarised in Table 1. This includes proven
or suggested assignments.
The DSC traces for both polymers do not display any clear
transitions, which suggests that they are amorphous. Glass
transitions are, however, often weak and may not always be
detected by DSC measurements. The TGA measurements reveal
The UV-vis spectroscopy of ortho-quinones, especially 9,10-
phenanthraquinone (2), has been studied many times.46–48 Fig. 4
displays the spectra obtained for 9,10-phenanthraquinone (2) in
chloroform obtained in the present work. The spectrum displays
4152 | J. Mater. Chem., 2009, 19, 4148–4156
This journal is ª The Royal Society of Chemistry 2009

View Article Online
Table 1 Summary of UV-vis spectra of compounds 2, 9 and 10 and
polymers 11–13 with suggested assignments
Band due
n–p* Banda to aggregationa
Compound/ p–p* Bandsa
Entry polymer
lmax/nm
lmax/nm
lmax/nm
1
2b
268 (29 000),
324 (4900),
418 (1700)
500 (sh)e
—
2
9b
334 (26 100),
na, 494 (2100)
301 (57 200),
na, 414 (23 500)
364 (37 000),
na, 493 (39 000)
309 (37 100),
na, 424 (22 700)
370 (51 600),
na, na
nd
525 (2100)
3
10b
11b
12b
13d
nd
nd
4
nd
526 (4100)
5
nd
nd
nd
6c
465 (4920)
a
Numbers in brackets are extinction coefficients measured for the
molꢂ1 cmꢂ1
na not
available. nd ¼ not detected. Determined for solution in chloroform.
corresponding transition, expressed in
L
;
¼
b
c
d
Data taken from reference 34.
Determined for solution in
tetrahydrofuran. e sh ¼ shoulder.
Fig. 3 SEC traces obtained for polymer 11 in THF (solid line) and
chloroform (dashed line). Marker is n-dodecane.
poor quinone moieties) whilst the other arises from an aggrega-
tion phenomenon. Fig. 5A shows the spectra of 9 in chloroform
and in toluene. As noted above, chloroform favours aggregation
of polymer 11, toluene does not. Consequently in the UV-vis
spectrum the intensity of the aggregation peak decreases on
going from chloroform solution to toluene. Further evidence of
this phenomenon is given by the fluorescence spectra in these
solvents. Thus, compound 9 does not display fluorescence in
chloroform, whereas it does in toluene solutions: see Fig. 5B.
This is consistent with the fact that aggregation can quench
fluorescence.
Fig. 4 shows the UV-vis spectra for solutions of polymers 11
and 12 in chloroform: Table 1, entries 4 and 5. The shapes of the
spectra are very similar to those of the corresponding models, the
transitions at 268, 324 and 418 nm. These can all be assigned to
p–p* transitions: see Table 1, entry 1.45–48 The n–p* is known to
be present around 500 nm,46 and can be observed as a shoulder
when a logarithmic scale is used.
The spectra of model compounds 9 and 10 are also shown in
Fig. 4 and summarised in Table 1, entries 2 and 3. These
compounds each display just two p–p* bands. If there is a third
band it is not apparent at >260 nm. In both cases the weak n–p*
band is believed to be hidden beneath the main bands. When
going from 10 to 9 it is interesting to note that both of the
observed transitions are shifted to longer wavelengths. This is
consistent with the more extended conjugation in the ‘‘linear’’
model 9 as compared to the ‘‘angular’’ 10. Thus, in 9 all the
linkages between the aromatic rings are para-linkages, whereas in
10 there are two meta- and two para-linkages. It is not clear why
the extinction coefficients for compound 10 are so much higher
than those for compound 9.
Finally on the UV-vis spectra of the model compounds, it is of
interest to consider the nature of the longest wavelength band of
compound 9, i.e. the band at 525 nm. This broad transition is
thought to be a combination of two peaks. One is the expected
long wavelength p–p* transition (the bathochromic shift
compared to 10 and 2 is believed to be due to a charge-transfer
phenomenon between the electron-rich fluorene and the electron-
Fig. 4 UV-vis spectra of 9,10-phenanthraquinones 2 (solid line), 9
(dashed line) and 10 (squares) and polymers 11 (dotted line) and 12
(triangles) for solutions in chloroform.
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 4148–4156 | 4153

View Article Online
Fig. 5 (A) UV-vis spectra of model 9 in chloroform (solid line) and in toluene (dotted line); (B) fluorescence spectrum of model 9 in toluene (lexcitation
:
350 nm).
main difference being that the peaks are red-shifted by up to
30 nm on going from the model compound to the corresponding
polymer. This is consistent with an increase in the length of the
conjugated system. As expected, this shift is larger in the case of
‘‘linear’’ polymer 11 and this is consistent with what has been
discussed above concerning the meta- and para-linkages. In other
words, the small shift observed when going from model 10 to
polymer 12 is due to the meta-linkages, which do not allow full
conjugation of the polymer backbone.5 In contrast, the longer
wavelength peaks (at 494 nm and 414 nm for 9 and 10, respec-
tively) are almost unchanged in the UV-vis spectra of polymers
11 and 12, compared to those of compounds 9 and 10. This
observation is consistent with the charge-transfer nature of the
493 nm transition in the case of 11 and to the lack of conjugation
typical of meta-aryls in the case of 12.
distorted the shape of the CVs of 9,10-anthraquinones, such as
the formation of dimers and the influence of trace amounts of
water or other protic donors, have similar effects in the case of
the ortho-quinones.47,48,51 These effects have been reported
previously.51
The results obtained for the cyclic voltammetry of models 9 and
10, together with 9,10-phenanthraquinone (2), for solutions in
dichloromethane are gathered in Table 2. The shapes of the CVs for
the two model compounds are very similar to what was observed
for compound 2. However, the gap between the cathodic and
anodic peaks of the first reduction step DE(I) was much larger in
the case of the models (550 and 570 mV) than for phenan-
thraquinone (2) (120 mV). This phenomenon could be related to
interactions between molecules occurring during the electro-
chemical processes, or to significant electronic and structural
reorganisation upon reduction. The latter is more likely because
the gap between the first and second reduction steps increases
significantly when going from 9,10-phenanthraquinone (2) to
both models: see Table 2. This increase corresponds to a stabili-
sation of the semiquinone state compared to the dianion state.
The shift in the half-wave potential when going from
9,10-phenanthraquinone (2) to both models is easier to ratio-
nalise. As for the parent anthraquinone-based derivatives,47 the
The UV-vis spectrum for polymer 13 as a solution in THF is
summarised in Table 1, entry 6.35 Assignments are those of the
group based in China. The spectrum differs significantly from
those of the present models 9 and 10 and polymers 11 and 12
measured for solutions in chloroform. The transition that is
present in the spectrum of 9 at 493 nm is shifted in the spectrum
of 13 to 465 nm. This is possibly due to the fact that chloroform,
a protic solvent, stabilises charge-transfer transitions from elec-
tron-rich aromatic systems to quinones (owing to the stronger
nucleophilic behaviour of their ketones).47 Given the intensity of
the band at 465 nm (3 ¼ 4920 L molꢂ1 cmꢂ1), it seems unlikely
that it is due to an n–p* transition.
Electrochemical properties
The electrochemical behaviour of ortho-quinones is very similar
to that of the more extensively studied para-quinones. Phenan-
thraquinone (2) is known to display two successive one-electron
reduction steps in aprotic solutions, similarly to what can be
observed for 9,10-anthraquinone (1).49,50 However, the difference
in resonance energy between the quinone and the dianion is
higher in the case of 9,10-anthraquinone (1),51 so the half-wave
potential of 9,10-phenanthraquinone (2) is shifted to more
positive potentials relative to 9,10-anthraquinone (1). Fig. 6
shows the CV of 9,10-phenanthraquinone (2) in dichloro-
methane. It can be seen that some of the phenomena that
Fig. 6 Cyclic voltammogram of 9,10-phenanthraquinone (2) (3 mmol
Lꢂ1) in dichloromethane; electrolyte: TBAPF6 (0.1 mol Lꢂ1); scan rate:
0.1 V sꢂ1; reference Ag/AgCl.
4154 | J. Mater. Chem., 2009, 19, 4148–4156
This journal is ª The Royal Society of Chemistry 2009

View Article Online
Table 2 Electrochemical data from cyclic voltammetry
Model/polymer
Ered1 (I)/mV
DE(I)/mV
Ered2(II)/mV
DE(II)/mV
EAd/eV
HOMOe/eV
LUMOe/eV
Eg CVf/eV
Eg Optg/eV
2a
ꢂ490
ꢂ610
ꢂ600
ꢂ550
ꢂ570
120
570
550
240
260
ꢂ1090
ꢂ1780
ꢂ1820
ꢂ1160
ꢂ880
400
200
9a
3.8
3.9
4.0
4.0
ꢂ6.2
ꢂ6.3
ꢂ6.2
ꢂ6.3
ꢂ4.2
ꢂ4.3
ꢂ4.3
ꢂ4.3
2
2
1.9
2
2.30
2.30
2.00
2.10
10b
11b
12b
500
c
240
a
Measurements performed in degassed DCM. b Measurements performed in acetonitrile. c Only the cathodic peak was observed. d Electron affinities
e
52,53
were calculated from the onset potentials for reduction Eonred according to the equation EA ¼ 4.4 + Eonred
.
HOMO and LUMO levels determined
Electrochemical band gap determined from the HOMO and LUMO levels. g Optical band gap determined from
the onset of adsorption. Electrolyte: TBAPF6, 0.1 mol Lꢂ1. Scan rate: 0.1 V sꢂ1. Reference: Ag/AgCl. (I) First reduction step; (II) second reduction step.
f
54
from the onset of reduction Eonset
.
strong electron-donor nature of the two aryl substituents in the
2,7- or 3,6-position has the effect of shifting the first reduction
peak cathodically.
diaryl-phenanthraquinones,35 this observation is difficult to
rationalise and could be due to the differences in the conditions
used to perform our measurements.
The CVs for polymers 11 and 12 were obtained for cast films
on a glassy carbon electrode. They were measured as films partly
due to their poor solubility in suitable solvents for CV studies
and partly because in an electronic device they would usually be
present as films. As is common for the CVs of polymer films, due
to slow diffusion effects, changes in solvation and aggregation of
the polymer chains, as well as disproportionation, the CVs of the
two polymers are quite distorted: see Fig. 7.36,47,55 Interestingly,
polymer 11 displays not the expected two reduction steps, but
three. This may be due to interactions between the phenan-
thraquinone units in aggregates. Polymer 12, however, does not
display any anodic peak for the second reduction step, which
suggests the occurrence of a side reaction.
The electron affinities (EAs) that were measured using
a ferrocene internal reference, for both polymers, are near 4.0 eV,
which lies amongst the highest electron affinities reported for
conjugated polymers. Table 2 displays the position of the
HOMO and LUMO levels for the different materials, as well as
their electrochemical and optical band gap. Given the high first
reduction potential measured for 9–12, the EAs measured and
the position of the LUMO level determined for these materials
are in good agreement with the results obtained for related
anthraquinone derivatives, which displayed EAs near 3.6 eV and
first reduction potential near ꢂ0.85 V (against Ag/AgCl).47,48
Notably, the electrochemical band gap closely matches the
optical band gap determined from the onset of adsorption of the
different materials. These EAs and LUMO levels range amongst
the highest reported in the literature: for example, benzimida-
zobenzophenanthroline ladder type polymer56 has a LUMO at
ꢂ4.4 eV (using ꢂ4.8 eV as the level of the SCE reference, as in the
present study) and fullerene C60 has an EA of 3.9–4.1 eV.57
The results obtained for the CVs of the two polymers are
summarised in Table 2. It is interesting to see that in each case the
first half-wave potential is shifted to slightly higher potentials,
compared to the models, suggesting that the effect of the fluorene
moieties on CT phenomena is split between two phenan-
thraquinone units. The reduction potential measured for poly-
mer 11 compares well with that reported in the literature for its
n-hexyl parent 13.35 However, the relative positions of the two
polymers’ first reduction peaks, with respect to that of 2, are
opposite. Indeed, the reduction of 13 is shifted anodically
compared to that of 2. As the electrochemical potential
measured by Hanif et al. was not placed in the context of other
Conclusions
The synthesis of copolymers using Suzuki couplings between
2,7-dibromophenanthraquinone (7) and fluorene derivative 4
and between 3,6-dibromophenanthraquinone (8) and fluorene
derivative 4 is described. Model compounds 9 and 10 were
Fig. 7 Cyclic voltammogram of polymer (A) 11 and (B) 12 films on glassy carbon electrode in acetonitrile; electrolyte: TBAPF6 (0.1 mol Lꢂ1); scan rate:
0.1 V sꢂ1; reference Ag/AgCl.
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 4148–4156 | 4155

View Article Online
20 R. Pacios and D. C. C. Bradley, Synth. Met., 2002, 127, 261.
21 D. Gebeyehu, C. J. Brabec and S. Sariciftci, Thin Solid Films, 2002,
403–404, 271.
22 X. L. Chen and S. A. Jenekhe, Macromolecules, 1997, 30, 1728;
F. E. Arnold and R. L. Van Deusen, Macromolecules, 1969, 2,
497.
23 G. Zubay, Biochemistry, Brown WC Publishers, Dubuque, Iowa,
1998, ch. 16, pp. 347–358.
24 G. Zubay, Biochemistry, Brown WC Publishers, Dubuque, Iowa,
1998, Chapter 17, pp. 377–382.
25 H. Etori, T. Kanbara and T. Yamamoto, Chem. Lett., 1994, 461.
26 T. Yamamoto and H. Etori, Macromolecules, 1995, 28, 3371.
27 P. Hodge, G. A. Power and M. A. Rabjohns, Chem. Commun., 1997,
73.
28 G. A. Power, P. Hodge, I. D. Clarke, M. A. Rabjohns and
I. Goodbody, Chem. Commun., 1998, 873.
29 Y. Muramatsu and T. Yamamoto, Polymer, 1999, 40, 6607.
30 Y. Muramatsu, T. Yamamoto, M. Hasegawa, T. Yagi and
H. Koinuma, Polymer, 2001, 42, 6673.
prepared similarly from the same quinones but in combination
with fluorene derivative 3. The structures of these new materials
were confirmed by FT-IR and 1H NMR spectroscopy. The UV-
vis spectroscopy of the polymers and their model compounds
clearly showed that a linear structure, such as that found in
polymer 11, favours aggregation. Aggregation is in part thought
to be due to the fact that the carbonyl groups of each quinone
moiety are pointing in the same direction. This results in the
phenanthraquinone moieties having significant dipoles. They are
expected to pack with neighbouring dipoles anti-parallel, as
shown in the X-ray crystal structures shown in Fig. 2. The anti-
parallel packing also reduces the steric interactions between
neighbouring molecules or polymer chains. Such aggregation
phenomenon, which indicates strong overlapping of the molec-
ular orbitals of adjacent molecules or polymer chains, together
with the high electron affinities measured for polymers 11 and 12,
suggests that materials of this general type may be useful for
organic electronics applications.
31 I. Goodbody, P. Hodge and A. Ben-Haida, React. Funct. Polym.,
2008, 68, 414.
32 P. Hodge and J. E. Gautrot, Polym. Int., 2009, 58, 261.
33 G. J. Gleicher, in The Chemistry of Quinonoid Compounds, ed. S.
Patai, Wiley, London, 1974, p. 18.
34 J. E. Gautrot, Synthesis of Novel Conjugated Polymers with High
electron Affinities for Applications in Electronic Devices, PhD
thesis, University of Manchester, 2004.
35 M. Hanif, P. Lu, M. Li, Y. Zheng, Z. Xie, Y. Ma, D. Li and J. Li,
Polym. Int., 2007, 56, 1507.
36 J. E. Gautrot and P. Hodge, Polymer, 2007, 48, 7065.
Acknowledgements
We thank Avecia for a PhD Studentship (JEG) and EPSRC for
financial support under the Carbon Based Electronics Pro-
gramme.
^
37 M. Belletete, S. Beaupre, J. Bouchard, P. Blondin, M. Leclerc and
G. Durocher, J. Phys. Chem. B, 2000, 104, 9118.
38 J. C. Miller, J. S. Meek and S. J. Strickler, J. Am. Chem. Soc., 1977,
99, 8175.
39 J. Schmidt and E. Junghans, Chem. Ber., 1904, 37, 3558.
40 M. V. Bhatt, Tetrahedron, 1964, 20, 803.
41 SMART (Version 5.625), SADABS (Version 2.03a) and SHELXTL
(Version 6.12). Bruker AXS Inc., Madison, Wisconsin, USA, 2001.
42 G. M. Sheldrick, SHELX97, Program for Crystal Structure Analysis
References
1 W. J. Feast, J. Tsibouklis, K. L. Power, L. Groenendal and
E. W. Meijer, Polymer, 1996, 37, 5015.
2 A. G. MacDiarmid, Angew. Chem., Int. Ed., 2001, 40, 2581.
3 J. Jagur-Grodzinski, Polym. Adv. Technol., 2002, 13, 615.
4 (a) W. J. Feast and R. H. Friend, J. Mater. Sci., 1990, 25, 3796; (b)
J. W. Y. Lam and B. Z. Tang, Acc. Chem. Res., 2005, 38, 745.
5 A. Kraft, A. C. Grimsdale and A. B. Holmes, Angew. Chem., Int. Ed.,
1998, 37, 403.
€
(Release 97-2), University of Gottingen, Germany, 1997.
43 H. Lee and R. G. Harvey, J. Org. Chem., 1986, 51, 2847.
44 K. Paruch, T. J. Katz, C. Incarvito, K.-C. Lam, B. Rhatigan and
A. L. Rheingold, J. Org. Chem., 2000, 65, 7602.
45 M. R. Rubin, J. Org. Chem., 1963, 28, 1949.
46 A. Kuboyama, F. Kobayashi and S. Morokuma, Bull. Chem. Soc.
Jpn., 1975, 48, 2145.
47 J. E. Gautrot, P. Hodge, D. Cupertino and M. Helliwell, New
J. Chem., 2007, 31, 1585.
48 J. E. Gautrot, P. Hodge, D. Cupertino and M. Helliwell, New
J. Chem., 2006, 30, 1801.
49 J. Q. Chambers, Electrochemistry of quinones, in The Chemistry of
Quinonoid Compounds, ed. S. Patai and Z. Rappoport, John Wiley
and Sons, Chichester, 1988, 719.
50 M. W. Lehman and D. H. Evans, J. Electroanal. Chem., 2001, 500,
12.
6 L.-X. Wang, X.-G. Li and Y.-L. Yang, React. Funct. Polym., 2001, 47,
125.
7 R. D. McCullough, Adv. Mater., 1998, 10, 93.
8 R. H. Friend, R. W. Gymer and A. B. Holmes, Nature, 1999, 397, 121.
9 U. Scherf and E. J. W. List, Adv. Mater., 2002, 14, 477.
10 H.-G. Nothofer, A. Meisel, T. Miteva, D. Neher, M. Forster, M. Oda,
G. Lieser, D. Sainova, A. Yasuda, D. Lupo, W. Knoll and U. Scherf,
Macromol. Symp., 2000, 154, 139.
11 L. Rothberg, Nature, 1990, 347, 518.
12 A. Pron and P. Rannou, Prog. Polym. Sci., 2001, 27, 135.
13 H. Fuchigami, A. Tsumura and H. Koezuka, Appl. Phys. Lett., 1993,
63, 1372.
51 N. Gupta and H. Linschitz, J. Am. Chem. Soc., 1997, 119, 6384.
52 S. Janietz, J. Barche, A. Wedel and D. Sainova, Macromol. Chem.
Phys., 2004, 205, 187.
53 S. Janietz, D. D. C. Bradley, M. Grell, C. Giebeler, M. Inbasekaran
and E. P. Woo, Appl. Phys. Lett., 1998, 73, 2453.
54 B. C. Thompson, Y.-G. Kim and J. R. Reynolds, Macromolecules,
2005, 38, 5359.
14 C. D. Dimitrakopoulos and D. J. Mascaro, IBM J. Res. Dev., 2001,
45, 11.
15 D. J. Sandman, Trends Polym. Sci. (Cambridge, UK), 1994, 2, 44.
16 F. Wurthner, Angew. Chem., Int. Ed., 2001, 40, 1037.
17 P. J. Brown, H. Sirringhaus and R. H. Friend, Synth. Met., 1999, 101,
557.
18 H. Sirringhaus, R. J. Wilson, R. H. Friend, M. Inbasekaran, W. Wu,
W. P. Woo, M. Grell and D. C. C. Bradley, Appl. Phys. Lett., 2000,
77, 406.
55 A. Iraqi, J. A. Crayston and J. C. Walton, J. Mater. Chem., 1998, 8,
31.
56 M. M. Alam and S. A. Jenekhe, Chem. Mater., 2004, 16, 4647.
57 P. Peumans and S. R. Forrest, Appl. Phys. Lett., 2001, 79, 126.
19 R. N. Marks, J. J. M. Halls, D. C. C. Bradley, R. H. Friend and
A. B. Holmes, J. Phys.: Condens. Matter, 1994, 6, 1379.
4156 | J. Mater. Chem., 2009, 19, 4148–4156
This journal is ª The Royal Society of Chemistry 2009